Plasmonics in graphene at infra-red frequencies

TL;DR

This paper investigates the properties of graphene plasmons at infrared frequencies, highlighting their potential for low-loss and highly localized wave propagation suitable for nanophotonics, with detailed analysis of loss mechanisms and frequency regimes.

Contribution

It provides a detailed theoretical analysis of plasmon losses in doped graphene at infrared frequencies, identifying conditions for low-loss propagation and potential applications in nanophotonics.

Findings

Graphene plasmons can have low losses below the optical phonon energy.

Losses increase due to electron-hole pair excitations in the interband regime.

A frequency bandwidth exists where phonon emission and electron-hole pairs contribute significantly to plasmon decay.

Abstract

We point out that plasmons in doped graphene simultaneously enable low-losses and significant wave localization for frequencies below that of the optical phonon branch $\hbar\omega_{Oph}\approx 0.2$ eV. Large plasmon losses occur in the interband regime (via excitation of electron-hole pairs), which can be pushed towards higher frequencies for higher doping values. For sufficiently large dopings, there is a bandwidth of frequencies from $\omega_{Oph}$ up to the interband threshold, where a plasmon decay channel via emission of an optical phonon together with an electron-hole pair is nonegligible. The calculation of losses is performed within the framework of a random-phase approximation and number conserving relaxation-time approximation. The measured DC relaxation-time serves as an input parameter characterizing collisions with impurities, whereas the contribution from optical phonons is estimated from the influence of the electron-phonon coupling on the optical conductivity. Optical properties of plasmons in graphene are in many relevant aspects similar to optical properties of surface plasmons propagating on dielectric-metal interface, which have been drawing a lot of interest lately because of their importance for nanophotonics. Therefore, the fact that plasmons in graphene could have low losses for certain frequencies makes them potentially interesting for nanophotonic applications.